As the game started at Elland Road the rain was falling. Some pundits expected Leeds to pressure Chelsea early on by employing the kind of fast tempo that the downpour might allow. The Yorkshire side did not disappoint in the opening minutes, easily controlling the first five or so.
In the opening half both teams were quick to recycle possession and looked professional in their approach. El Hadji Diouf stood out, despite the fact that he has struggled to score goals this year. His commitment to close down Chelsea at the back was impressive and one block even got applause from his teammate Michael Brown. As the first half developed it became a battle for midfield supremacy with Frank Lampard looking motivated despite not having seen much football recently. Chelsea were unlucky not to be awarded a penalty when Victor Moses was clipped on 23 minutes.
For all the passion shown by Leeds, they were playing too narrow a game and did not see much of the ball in the final third. As a result the Chelsea defense was not stretched. Similarly, Leeds looked very comfortable in defense with the back four well protected by midfield Michaels Tonge and Brown, while Chelsea also saw little of the ball in the final third. Fernando Torres in particular had another poor half and failed to hold the ball up or link with other players. He seemed intent on trying to beat the Leeds defense on his own and gave the ball away too often.
Chelsea conceded a goal on 37 minutes, Leeds’ first attempt. If any game highlights the inability of David Luiz to be a world-class player, this one did, as he ran the ball towards the Leeds goal, losing possession. It was then a simple matter for the fleet-footed Jerome Thomas to exploit the gap at the heart of the Chelsea defense, resulting in a clinical finish by Luciano Becchio.
After the goal the Sky Sports commentator remarked that Leeds were now in the driving seat despite Chelsea having been superior in every almost department.
Indeed, looking at the raw data for the first half makes it clear that Chelsea were the more dangerous team and that the half-time scoreline flattered Leeds, despite their defensive hard work.
At the start of the second half Leeds paid the price and sat too deep and invited Chelsea to attack them. An apparent lack of concentration on the part of the Leeds ‘keeper let Chelsea back into the game. Leeds reacted very well to conceding such an early second-half goal and the game opened up, with neither side controlling for the first 20 minutes or so. Tonge for Leeds drove forward in this period, yet Chelsea’s Lampard, who I would make man of the match, exerted great influence in the midfield.
fatigue soon took its toll on Leeds and, in particular, Thomas looked tired and was taken off. The midfield that had protected the defense so well went missing for Chelsea’s second and third goals. At this point the game was effectively over and Chelsea took control of possession and slowed the play down . Chelsea were efficient and it seemed cruel that Leeds, who had played with such energy, inevitably ran out of steam.
The delivery data is for crosses and long-range shots and shows how narrow a game is (more deliveries and more expectation of goals) but we must consider if a delivery is played to five attackers in the box or just one.
A half chance is as it sounds: a shot or header that we do not expect to be scored; a chance is one step removed from a near miss, where there is danger but no expectation of a goal; a near miss is we expect the ball to hit the target but doesn’t.
You will note that I did not key near misses in the game but the chance to Torres on 59 minutes could almost be considered a near miss as he had the goal at his mercy. I only gave a chance here as there was a Leeds defender attempting to stop him.
The reason why shot on target data is collated is to look for patterns in order to define the relative expectation of a goal.
In general you will see clusters of goal attempts followed by a goal.
My challenge to you is to start doing this in games and see if you can predict goal expectation.